基于FFRLS和EKF算法的磷酸铁锂电池SOC估算研究

Research on SOC Estimation of LiFePO4 Battery based on FFRLS and EKF Algorithm

摘要:

准确估计荷电状态是磷酸铁锂电池安全工作和电动汽车正常使用的基础，以混合的简化电化学模型为基础，使用遗忘因子递推最小二乘法(FFRLS)对模型的参数进行辨识。对安时积分法进行参数修正，降低了充放电倍率，温度等因素的影响，并对扩展卡尔曼滤波(EKF)进行改进，使得观测值的修正能力提高。以参数修正过的按时积分法的方程为状态方程，结合开路电压法，利用改进过的EKF进行SOC的估计。与安时积分法相比，SOC的估算效果提高了不少，使估算偏差保持在3%以内。

Abstract:

Accurately estimating the state of charge is the basis for the safety work of LiFePO4 battery and the normal use of electric vehicles. Based on a simplified electrochemical model, the parameters of the model are identified by using the Forgetting Factor Recursive Least Squares (FFRLS) method. The parameters of Ampere-hour integration method are revised to reduce the influence of factors such as charge-discharge rate and temperature, and the EKF is improved to improve the correction ability of the observed values. The Ampere-time integration equation where parameter is modified is the state equation, combining with the open-circuit voltage method, and the EKF is used to estimate the SOC. Compared with the ampere-hour integration method, the estimation effect of SOC is much improved, and the estimation deviation is kept within 3%.